I was reading left-leaning blog Counterspin Central when I came across a reference to an article which indicated that a non-race-oriented alternative to affirmative action, the Texas Top 10% Plan under which top high school students are guaranteed admission to Texas universities is failing. Always trying to learn more and remain open-minded, I read the article.

After reading past the first paragraphs, containing the inevitable emotive "personal angle" spin, and arriving at "the meat," I started to see some real problems...

'Faltering' Admissions?

First, the article characterizes the 10% policy as "faltering", calling such admission programs "embattled", and implying it was failing as an alternative to affirmative action. The basis for the charge of 'faltering' is the 1% decline in African-American enrollment from 2000 to 2001 depicted to the right. Note a similar dip from 1999 to 2000 for Hispanics: we could have said the policy was "faltering" that year too, yet here we see a slight increase again.

Further, when we look at non-African-Americans enrollment, or even add the two together, we see no decline. The reporter's charge of "faltering" is only true (from this data) when the we equate "minority" with African-Americans.

Observe the pre-1997 admissions decline among African-Americans. What policy was in effect at this time? Affirmative action. Its clear that African-American enrollment had already been steadily declining since 1994, for reasons not explored within the article. (The joker in me asks Did you also write a 1996 article about affirmative action 'faltering'?) If there are other factors which could cause a decline under affirmative action, is it not also possible those factors could be contributing here as well? (In fact, I found some, and describe them below.)

UT Law School Data

Below that chart, we are supplied with another showing decling numbers of African Americans in UT Law School. Here again, questions arise: Why aren't pre-1997 numbers available so we can see what the most recent trend was under affirmative action?

And note carefully that what we're seeing here is currently attending students, not admissions. The impact of a dip in admissions like the one seen in the previous chart between 1994 and 1998 would be spread out over four years attendance. Further, law schools admit those who matriculate from undergraduate programs, so a change in undergraduate enrollment could take years to have an impact on law schools.

I did a quick Google search which confirmed that this year's (2002) UT Law's African-American attendance increased 31% (compared to the previous year), putting this year's African-American attendance at 4% -- which would have been the second-highest bar on the chart had the number not been omitted from the results presented here. The omission of the most recent data from the chart is particularly supicious given that the reporter has already cited it elsewhere in her article.

Even the data in the article itself doesn't support the caption given to the chart!

Getting the Whole Picture

Frustrated by data missing from the article, and suspicious about the switch from enrollment numbers to attendance numbers, I went in search of the raw data. A few clicks around the UT web site produced this summary [PDF] which reveals the actual UT overall minority participation levels since 1996:

Overall, minority participation since 1996 (last affirmative action year) is signficantly up, not down. There's a 1% decline in some number's from last year's peak. Accoring to the analysis, "This was due to a slight decrease in the percentage of Asian Americans" (emphasis mine), and that African American and Hispanic populations have remained steady.

The summary goes on to say [Hopwood is the 1996 case banning affirmative action]:

Generally pre-Hopwood diversity levels were restored by 1998 or 1999 in the admitted and enrolled populations and have held steady. Since 1996, among the enrolled, Whites experienced the greatest decrease (-3%); Asian-Americans showed the greatest increase (+3%); while African-Americans and Hispanics fluctuated slightly (+/-1%).

Paints a very different picture, doesn't it?

Even ignoring the 2002 data (as the reporter apparently did), the only way you could paint 2001's minority participation levels as "faltering" would be, as I noted above, to equate "minority" with African-American. Otherwise, overall minority participation sharply increased in every participation category that year!

The article was somewhat more accurate about the UT Law School situation. Minority participation rates are, as asserted, currently lower than before affirmative action was abolished. But again, the central contention of the article and chart -- falling minority participation in the law school -- is further put to lie by the actual post-Hopwood statistics which depicts instead a steady rise in minority admissions:

Whether the Texas ten percent plan can achieve parity with a system of quotas and differing admissions standards by race at the graduate level (14-20.5%) is not yet known. Nor am I able to find pre-Hopwood enrollment statistics. Nonetheless, this data again demonstrates a trend which is the exact opposite of the picture painted by the "Decline in law school" graph!

(Lies, damned lies, and statistics!)

Other Factors

Another quick Google search revealed that this lower law school minority admission rate, and the dips after the enactment of the 1997 law were not due to universities accepting a lower percentage of minority applicants, but rather because fewer minorities chose to apply during those years. (For example, UT Law experienced a 42% drop in African-American applicants.) So some of the decrease in minority enrollment to these schools was minority-imposed, not as a result of otherwise-qualified applicants failing to be admitted by a new policy. Said one administrator: "Some of our competitors are bad-mouthing us, telling minorities not to come to Texas because they are not welcome." (In fact, both Time and Hispanic Business Magazine ranked UT Law as #1 in minority outreach.)

Another factor contributing to this intial dip appears to have been an initial failure to offer adequate financial aid to top 10% students, which has since been alleviated with a $4,000-per-year stipend to qualfied individuals. The scholarship was mentioned in the article, but the reporter did not explain its role in the dip in minority participation, as this quote does:

"This modest decline is far less than what many had predicted in the wake of last year's Hopwood vs. University of Texas Law School decision," said Ray M. Bowen, president of Texas A & M. "It illustrates that our admissions staff has done an excellent job generating interest in Texas A & M University." Bowen said the figures support an initial conviction that Hopwood created a financial-aid problem, not an application or admissions problem.

Since the article talks only about the removal of affirmative action, the reader is left with the impression there are no other factors which could explain this dip in enrollment. Important data has again been ommited.

How to Measure Success?

The goal of affirmative action programs, as I understand them, is (a) to increase campus diversity, and (b) to given a necessary boost to disadvantaged minories desiring a college education.

The article measures success by "diversity", and diversity only by racial percentages. Important cultural differences between wealthy and poor minorites are ignored: the son of a wealthy minority lawyer who attends an elite prep school increases this measure of campus "diversity" just as much as a poor minority immigrant from a rough neighborhood.

My research indicated the current policy may be meeting the above goals better than its predecessor when socioeconomic status and geographic location are also considered in a measure of diversity:

One early finding is that under the new law, the University of Texas is drawing students from a greater number of high schools and from a wider geographic area ... High schools that began sending students after sending none or few in the past were principally inner-city minority schools in Dallas-Fort Worth, Houston and San Antonio and rural white high schools in east and northeast Texas.

While I understand the press's tendancy to just look at stastics to draw their conclusion, it would have been nice if the reporter would have shown a little balance by including this information

Muddled Thinking

Several "persuasive" arguments are set forth in this article, argued from the reporter's point of view. I found the logic in these to be flawed. For example:

The plan also has the dubious distinction of successfully recruiting minority students only because it relies on the existence of racially and ethnically segregated schools.

Since the program draws on the top 10% of each school, this true -- it attempts to remedy problems caused by poorly-funded minority schools. But so allegedly does affirmative action (though some charge it actually disproportionately benefits upper-middle class and wealthy recipients). Does that make affirmative action 'dubious' as well?

Further, what if these "segregated" schools didn't exist? Isn't the prevailing wisdom that minorities placed in integrated, equally-funded environments will do equally well? Then what harm would the top ten percent program be, since minorities would have achieved parity with whites? It would still provide an incentive to motivate high school students of all colors and income levels to academic excellence.

Another argument the reporter advanced regards the alleged hidden 'complexity' of the plan. But the complexity alluded to was not in the ten percent admission plan, but rather in one specific high school's qualification for making the "top ten percent":

For Joey Delgado, a student in an affluent, academically competitive high school, the law is unfair. Despite getting almost all As in Advanced Placement courses and a 1360 on the SAT, it still wasn't good enough. To get in the top 10 percent at Westlake High School, students not only have to get perfect grades, but get those grades in the toughest classes.

Once again, proof by anecdote, and an obvious contradiction to boot. The state law is unfair because a student feels he's getting a bum rap from his local high school. (This is like saying state speeding laws are unfair because a specific cop pulled you over for speeding when you swore you weren't.)

Note Joey only got "almost all" As on his AP courses. How many did he take? Three? Ten? What did he get on the ones below "A" level? And what other non-AP courses did he take, for what grades? And of course, the SAT is irrevelant to GPA, but that's mentioned anyway, for emotional impact. There's not enough data here to agree with the reporter, and further, the omission of this data while the ten percent program is pronounced "unfair" hints at an axe being ground.

(Probably, his high school has some criteria to ensure that students getting A's in tough AP courses aren't eclipsed academically by students who excel at volleyball and basket weaving; these rules probably exist because its the alternative which would be truly unfair. Regardless, it's his school's policy we're discussing, not state law. His absense from his school's top ten percent will affect other scholarships too, such as National Merit. I'm sure those must be 'unfair' on the same basis. Perhaps this reporter dislikes meritocracies?)

The "complexity" mentioned in the title of that section of the article is not directly explained. Presumably the program is complex for Joey because he must get (we are taking the reporter's word here) "perfect grades" (e.g. straight A) in AP classes. I don't know how other readers will react, but my reaction was: If a student found that too difficult to understand, they should probably not be in the top ten percent of their graduating class. Difficult to achieve perhaps, but not complicated. (Perhaps it was the reporter, not the student, who had trouble understanding this rule.)

Sometimes, the reporter can't seem to figure out which statistics to align with what points. For example, she starts off saying:

Critics of the state's 10 percent plan say any modest gains in minority enrollment are a result of larger increases in the Texas college-age minority population...

And then continues by supplying the overall minority population percentages, not the change in population over these years:

Overall, the state's population is 33 percent Hispanic and 11.5 percent black, and college-age population is 40 percent Hispanic and 12.3 percent black, according to a Princeton study.

This information is irrevelant to the charge that any 'modest' gains were due to an increase in Texas's college-age minority population.

Pulling Our Emotional Chain

No-one in the article fully defends the top ten percent program. The only student quoted as voicing support ("a white sophomore who did not graduate in the top of his high school class") comes off sounding racist, as well as academically-challenged:

"It has its flaws, but it is better than allowing race to be considered," said David Snider, a white sophomore who did not graduate in the top of his high school class. "It's a lot better than filling a requirement to let a certain number of minority students in."

We wouldn't want to let a certain number of minority students in, the implication seems to be. The focus of this quote is on "letting in" minority students -- hinting at the bad old days of southern racism and segregation.

When a reporter chooses a quote like this, there are exactly two possible reasons why. One is that the quote really was the only one obtained, chosen at random, and thus could be a reasonable sample of the prevailing attitude among individuals with the same general view. The other possibility is out of thousands of students and hundreds of administrators, the reporter was able to obtain quite a number of quotes, and carefully chose the one which connoted the impression she wished to create.

Unrelated sidebars and photos decorate the article. The choice of these subject matter here also gives an indication of what the author or editor was thinking, or hoping to make us think.

At the top, UT students gather to protest eggs being thrown at a statue of MLK. Interesting that an incident implying racism among Texas students (well, at least by one student with some eggs) was chosen to accompany an article on the state's "top ten percent" policy. Are we to conclude the banning of affirmative action five years ago caused the individual or group to do this last week, despite current similar minority enrollment levels?

(I'm reminded again of the administrator's quote: "Some of our competitors are bad-mouthing us, telling minorities not to come to Texas because they are not welcome." That certainly seems to be the image communicated by the choice of this photo.)

Below the protest, a survey question is posed. Again, only critics of the program are cited, with no counter-argument:

Texas has had mixed results with its race-neutral policy of admitting the top 10 percent of every high school class to any public state university. Critics argue that the Texas system doesn't consider test scores, essays, grades or other factors. Overall, which admissions program do you think is better?

Incidentally, the critics not-cited here are quite wrong on each count. The UT admission criteria actually expanded after affirmative action. Under affirmative action, admissions decisions relied only on "the combined SAT score (or a concorded ACT score) and high school class rank" to predict freshmen scores. Affirmative action was then used to admit the best in each racial category.

The admissions criteria expanded to include other criteria after banning affirmative action. Current criteria include all of the above (except race, of course), and add leadership, awards and honors, work experience, service to school or community, scores on two separate essay questions, and a myriad of other "special circumstances" including income, average SAT/ACT against school's average, and family circumstances -- exactly the opposite of the critics' allegations. Sadly, the reporter doesn't include these facts.

Finally, at the bottom we see a picture of an elementary school (elementary school???) teacher participating in an anti-war protest on MLK day. The connection? A single "affirmative action" sign admist all the anti-war signs. Again, while the connection between this elementary school teacher and a state college-level scholarship program is tenuous, but the emotional implication is clear: Caring, anti-war people are against this program.

Conclusion

The tone of the article, overall, is dismissive. Important data is omitted, personal anecdotes appear tailored to the reporter's views, and accompanying material seems to be chosen to present an image of student body tainted with racism. The thesis of this article is only supportable in a very narrow interpretation of the data -- a one-year, 1% enrollment decline among the second-largest minority. The star exhibit -- the UT Law School -- proves the opposite, and a supporting graph was only made possible by switching from admission to attendance figures and omitting the most recent data. I can't see that a typical person would come to this conclusion when presented with all the data, sans rhetoric.

When data excluded from the article is considered, the reporter's contention is further weakened: Not only does it appear that the ten percent program and current admission policies have generally produced diversity comparable to that under affirmative action, but law school minority admissions and enrollment are shown to be rising, not falling.

Only one broad assertion in the article, that law school minority participation is still lower than pre-1996 levels, stands, though the omitted data shows even this level appears to be trending briskly back towards its pre-1996 levels.

A non-reporter (me) who knew almost nothing about the topic earlier today, in several hours has written a summary pointing several major problems with the article, and uncovered important omitted data. Why do these flaws exist? Either incompetance or bias. From the tone of several of the reporter's comments, I'd suspect the latter.

Since the reporter appears biased, and since bad PR is said to be one of the main problems hurting the Texas 10% program, I can't help but wonder if this article isn't also meant to dissuade potential students from applying to UT, and/or paint Texas in a poor light to dissuade those who might consider similar policies for Michigan.

(Why did I write this article?)